Am 15.10.25 um 08:36 schrieb lirongqing:
From: Li RongQing lirongqing@baidu.com
Currently, when 'hung_task_panic' is enabled, the kernel panics immediately upon detecting the first hung task. However, some hung tasks are transient and allow system recovery, while persistent hangs should trigger a panic when accumulating beyond a threshold.
Extend the 'hung_task_panic' sysctl to accept a threshold value specifying the number of hung tasks that must be detected before triggering a kernel panic. This provides finer control for environments where transient hangs may occur but persistent hangs
should be fatal.
The sysctl now accepts:
- 0: don't panic (maintains original behavior)
- 1: panic on first hung task (maintains original behavior)
- N > 1: panic after N hung tasks are detected in a single scan
This maintains backward compatibility while providing flexibility for different hang scenarios.
Signed-off-by: Li RongQing lirongqing@baidu.com Cc: Andrew Jeffery andrew@codeconstruct.com.au Cc: Anshuman Khandual anshuman.khandual@arm.com Cc: Arnd Bergmann arnd@arndb.de Cc: David Hildenbrand david@redhat.com Cc: Florian Wesphal fw@strlen.de Cc: Jakub Kacinski kuba@kernel.org Cc: Jason A. Donenfeld jason@zx2c4.com Cc: Joel Granados joel.granados@kernel.org Cc: Joel Stanley joel@jms.id.au Cc: Jonathan Corbet corbet@lwn.net Cc: Kees Cook kees@kernel.org Cc: Lance Yang lance.yang@linux.dev Cc: Liam Howlett liam.howlett@oracle.com Cc: Lorenzo Stoakes lorenzo.stoakes@oracle.com Cc: "Masami Hiramatsu (Google)" mhiramat@kernel.org Cc: "Paul E . McKenney" paulmck@kernel.org Cc: Pawan Gupta pawan.kumar.gupta@linux.intel.com Cc: Petr Mladek pmladek@suse.com Cc: Phil Auld pauld@redhat.com Cc: Randy Dunlap rdunlap@infradead.org Cc: Russell King linux@armlinux.org.uk Cc: Shuah Khan shuah@kernel.org Cc: Simon Horman horms@kernel.org Cc: Stanislav Fomichev sdf@fomichev.me Cc: Steven Rostedt rostedt@goodmis.org
diff with v3: comments modification, suggested by Lance, Masami, Randy and Petr diff with v2: do not add a new sysctl, extend hung_task_panic, suggested by Kees Cook
Documentation/admin-guide/kernel-parameters.txt | 20
+++++++++++++-------
Documentation/admin-guide/sysctl/kernel.rst | 9 +++++---- arch/arm/configs/aspeed_g5_defconfig | 2 +- kernel/configs/debug.config | 2 +- kernel/hung_task.c | 15
++++++++++-----
lib/Kconfig.debug | 9
+++++----
tools/testing/selftests/wireguard/qemu/kernel.config | 2 +- 7 files changed, 36 insertions(+), 23 deletions(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index a51ab46..492f0bc 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -1992,14 +1992,20 @@ the added memory block itself do not be affected.
hung_task_panic=
[KNL] Should the hung task detector generate panics.Format: 0 | 1
[KNL] Number of hung tasks to trigger kernel panic.Format: <int>When set to a non-zero value, a kernel panic will be triggeredif
the number of detected hung tasks reaches this value.0: don't panic1: panic immediately on first hung taskN: panic after N hung tasks are detected in a single scan
A value of 1 instructs the kernel to panic when ahung task is detected. The default value is controlledby the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-timeoption. The value selected by this boot parameter canbe changed later by the kernel.hung_task_panic sysctl.
The default value is controlled by theCONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time option.The value
selected by this boot parameter can be changed later by thekernel.hung_task_panic sysctl.hvc_iucv= [S390] Number of z/VM IUCV hypervisor console
(HVC)
terminal devices. Valid values: 0..8 diff --gita/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst index f3ee807..0065a55 100644 --- a/Documentation/admin-guide/sysctl/kernel.rst +++ b/Documentation/admin-guide/sysctl/kernel.rst @@ -397,13 +397,14 @@ a hung task is detected. hung_task_panic ===============
-Controls the kernel's behavior when a hung task is detected. +When set to a non-zero value, a kernel panic will be triggered if the +number of hung tasks found during a single scan reaches this value. This file shows up if ``CONFIG_DETECT_HUNG_TASK`` is enabled.
-= ================================================= += ======================================================= 0 Continue operation. This is the default behavior. -1 Panic immediately. -= ================================================= +N Panic when N hung tasks are found during a single scan. += =======================================================
hung_task_check_count
[…]
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 3034e294..3976c90 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1258,12 +1258,13 @@ config DEFAULT_HUNG_TASK_TIMEOUT Keeping the default should be fine in most cases.
config BOOTPARAM_HUNG_TASK_PANIC
- bool "Panic (Reboot) On Hung Tasks"
- int "Number of hung tasks to trigger kernel panic" depends on DETECT_HUNG_TASK
- default 0 help
Say Y here to enable the kernel to panic on "hung tasks",which are bugs that cause the kernel to leave a task stuckin uninterruptible "D" state.
When set to a non-zero value, a kernel panic will be triggeredif the number of hung tasks found during a single scan reachesthis value.The panic can be used in combination with panic_timeout, to cause the system to reboot automatically after a
Why not leave the sentence about the uninterruptible "D" state in there?
This seem to say a kernel bug to cause hung task, but it maybe hardware failure(or virtio backend bug); so I do not keep it
Also, it sounds like, some are actually using this in production. Maybe it should be moved out of `Kconfig.debug` too?
I think hung task panic is a useful feature, it should move out of Kconfig.debug
Thanks
-Li
Kind regards,
Paul